Due to low stability, polypeptides are generally apt to denature and be degraded by proteinases and lose their activity. On the other hand, peptides are relatively small in size so that they are readily excreted through the kidney.
In order to maintain desired blood level concentrations and titers thereof, thus, protein medicines comprising polypeptides or peptides as active ingredients need to be frequently administered. However, because protein medicines are, for the most part, in a form suitable for injections, the maintenance of physiologically active polypeptides or peptides at appropriate blood levels requires frequent injections, causing significant pain to the patient. In order to overcome these problems, attempts have been made to provide maximum medicinal effects by increasing the stability of protein medicines in the blood and by maintaining blood medicine levels high for a long period of time. These long-lasting protein medicine agents are required not only to increase the stability of the protein medicines and maintain sufficient titers of the medicines themselves, but also to not cause immune responses in the patients.
Conventionally, highly soluble polymers such as polyethyleneglycol (PEG) are chemically grafted onto the surface of proteins with the aim of stabilizing the proteins, preventing the contact of proteinases with the proteins, and suppressing the renal loss of small-size peptides. Grafted to specific or a variety of different sites on proteins, PEG is useful for the stabilization and hydrolysis prevention of proteins without creating noteworthy side effects. In addition, grafted PEG increases the molecular weight of the proteins, thereby restraining renal loss of the proteins and maintaining the physiological activity of the proteins.
For example, WO 2006/076471 describes the use of B-type natriuretic peptide (BNP) in the treatment of congestive heart failure. BNP binds to natriuretic peptide receptor A (NPR-A) to trigger the synthesis of cGMP, thereby reducing arterial blood pressures. When PEGylated, BNP is described as elongating the physiological activity thereof for a long period of time. U.S. Pat. No. 6,924,264 also discloses an increase in the active period of exendin-4 by grafting PEG onto a lysine residue.
In order to increase the physiological activity thereof, a medicinal polypeptide is linked to both of the terminals of PEG to form a bis-conjugate (U.S. Pat. No. 5,738,846). On the other hand, two different medicinal proteins are linked to respective terminals of PEG to form a protein complex which have two different physiological activities (WO 92/16221). However, no significance was found in these protein drugs in terms of activity maintenance.
Also, it was reported that a fusion protein in which G-CSF and human albumin were linked to one PEG increased in stability (Kinstler et al., Pharmaceutical Research 12(12): 1883-1888, 1995). However, the modified drug with a G-CSF-PEG-albumin structure was found to increase in residence time by only about four times, compared to natural drugs alone, and to be only slightly increased in serum half life. Thus, the modified drug is not practically applied as a lasting agent.
When coupled with PEG, peptides become so stable as to extend the persistence thereof in vivo. However, when given high molecular weights, PEG makes the titer of the physiological active peptide significantly low and decreases in reactivity with peptides, resulting in a low yield.
An alternative for increasing the stability of physiologically active proteins in vivo takes recourse to gene recombination. A gene coding for a protein highly stable in the blood is linked to a gene coding for a physiologically active protein of interest, followed by transformation into animal cells which are then cultured to produce a fusion protein.
For example, a fusion protein in which albumin or a fragment thereof, known to be the most effective in stabilizing proteins thus far, is fused to a physiological active protein of interest has been reported (WO 93/15199 and 93/15200, EP Publication No. 413,622). Also, a fusion protein of interferon alpha and albumin, produced from yeast by Human Genome Sciences (trade name: Albuferon™) increased its serum half-life from 5 hrs to 93 hrs, but suffers from the critical disadvantage of being decreased in bioactivity to less than 5% of that of naive interferon (Osborn et al., J. Phar. Exp. Ther. 303(2): 540-548, 2002).
As for peptides, their modifications are mentioned in WO 02/46227 which discloses that GLP-1, exendin-4 and analogs thereof are fused to human serum albumin or immunoglobulin fragments (Fc) using genetic recombination techniques and in U.S. Pat. No. 6,756,480 which discloses fusion proteins of parathyroid hormone (PTH) or analogs thereof and immunoglobulin fragments (Fc). These approaches can overcome low pegylation yield and non-specificity, but are disadvantageous in that serum half life is not significantly increased and in some cases, low titers result. Various peptide linkers are used to maximally increase serum half life, but show the high possibility of causing immune responses. When given, a peptide having a disulfide bond, such as BNP, is highly likely to induce misfolding and thus is difficult to apply.
Other various fusion proteins are also known to be prepared by linking the Fc domain of immunoglobulin to interferon (Korean Patent Publication No. 2003-9464), interleukin-4 receptor, interleukin-7 receptor or erythropoietin receptor (Korean Patent No. 249572) through genetic recombination. PCT Patent Publication No. WO 01/03737 discloses a fusion protein in which a cytokine or a growth factor is linked through an oligopeptide linker to an Fc fragment of immunoglobulin. U.S. Pat. No. 5,116,964 describes LHR (lymphocyte cell surface glycoprotein) or CD4 protein which is fused to the amino or carboxy end of an immunoglobulin Fc domain using a genetic recombination technique. Also, U.S. Pat. No. 5,349,053 discloses a fusion protein in which IL-2 is linked to an immunoglobulin Fc domain. Many other Fc fusion proteins constructed using genetic recombination techniques are disclosed, examples of which include a fusion protein of an immunoglobulin Fc domain with interferon-beta or a derivative thereof (PCT Patent Publication No. WO 00/23472), an immunoglobulin Fc domain with an IL-5 receptor (U.S. Pat. No. 5,712,121), an immunoglobulin G4 Fc domain with interferon alpha (U.S. Pat. No. 5,723,125) and an immunoglobulin G2 Fc domain with a CD4 protein (U.S. Pat. No. 6,451,313). On the other hand, U.S. Pat. No. 5,605,690 teaches the use of a modified immunoglobulin Fc domain in the production of fusion proteins. For example, immunoglobulin Fc with amino acid residues modified particularly to complement binding sites or receptor binding sites is used to produce a TNFR-IgG1 Fc fusion protein using a genetic recombination method. Other fusion proteins of the modified immunoglobulin Fc domain which are produced using gene recombination techniques are disclosed in U.S. Pat. Nos. 6,277,375, 6,410,008 and 6,444,792.
Immunoglobulins function as antibodies, showing antibody-dependent cell cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC) and the sugar chains present in the immunoglobulin Fc domain are reported to play an important role in ADCC and CDC (Burton D., Molec. Immun. 22, 161-206, 1985). Immunoglobulins themselves, when free of sugar chains, are known to be similar in serum half life to the immunoglobulins having sugar chains, but to have a 10- to 1000-fold decrease in complement binding force and receptor binding force (Waldmann H., Eur. J. Immunol. 23, 403-411, 1993; Morrison S., J. Immunol. 143, 2595-2601, 1989).
U.S. Pat. No. 6,660,843 discloses the fusion of an Fc domain with a peptide of interest through a linker and the production of the fusion protein in E. coli using a gene recombination technique. For use in preparing complexes, a linker allows the selection of the conjugation sites between two proteins of interest and the orientations thereof, and enables the production of complexes in the form of homogenous or heterogenous monomers, dimers or multimers. In addition, when using this method, the complexes can be produced at a lower cost than when using mammal cells. In addition, the complexes may be produced in sugar chain-free forms. Because of the concomitant production of both the protein of interest and the immunoglobulin Fc domain in E. coli, however, this method is difficult to apply to a target protein when the native form of the target protein has a sugar chain. Taking advantage of inclusion bodies, this method is highly apt to induce misfolding. In the Fc fusion proteins produced using the genetic recombination techniques, fusion is possible only at specific sites, that is, an amino or carboxy terminus of the immunoglobulin Fc domain. The Fc fusion proteins are expressed only in homogenous dimeric forms, but not in monomeric forms. Further, fusion is possible only between glycosylated proteins or between aglycosylated proteins, but impossible between glycosylated proteins and aglycosylated proteins. If present, an amino acid sequence newly formed as a result of the fusion may induce an immune response. Moreover, the linker may be sensitive to enzymatic degradation.
In the development of fusion proteins using immunoglobulin Fc domains, nowhere has an attempt been made to give complexes of target proteins with human native Fc through a crosslinker in previous reports. Immunoglobulin Fc domains can be produced in mammal cells or E. coli using genetic recombination techniques, but nowhere has an attempt been made to produce only native immunoglobulin Fc domains free of target proteins at high yield and to apply them to lasting forms in previous reports. In addition, no attempts have been made to produce complexes of the recombinant immunoglobulin Fc with target proteins through crosslinkers.
As such, a variety of different methods have been performed in order to conjugate physiologically active polypeptides with polymers. In conventional methods, polypeptides can be improved in stability, but with significant reduction in activity, or can be improved in activity irrespective of stability. Therefore, there is still a need for a method that can increase the stability of protein medicines with a minimum reduction in modification-induced activity.
In this context, the present inventors developed a protein complex which is improved in serum half life with high activity by linking an immunoglobulin and a physiologically active polypeptide respectively to opposite termini of a non-peptidyl linker as disclosed in Korean Patent Nos. 10-0725315 and 10-0775343, which are incorporated by reference in their entirety.
A protein complex in which an immunoglobulin and a physiologically active polypeptide are respectively linked to opposite termini of a non-peptidyl polymer is conventionally prepared by linking a non-peptidyl polymer preferentially with a physiologically active polypeptide and then with an immunoglobulin Fc domain. However, this conventional preparation method produces lots of undesirable impurities, as well, resulting in losing a large quantity of the physiologically active polypeptide. That is, the conventional method is economically unfavorable upon the industrial application thereof and the resulting complex must be purified in a somewhat complicated manner. In the case where the physiologically active polypeptide is in the form of a dimer, it produces a bridge form with a non-peptidyl polymer at both termini so that it cannot complex with an immunoglobulin Fc or can complex but at very low yield. On the other hand, when an immunoglobulin Fc domain is first linked to a non-peptidyl polymer, similar problems occur as well. Because an immunoglobulin Fc is a homo-dimer with two N-termini in close vicinity to each other, respective links are formed between the two N-termini of the immunoglobulin Fc and the opposite termini of the non-peptidyl polymer to produce a bridge form, so that no functional ends remain to be reacted with the physiologically active polypeptide. Accordingly, the production yield significantly decreases.